This document discusses the development of a photoplethysmography-based system for detecting atrial fibrillation (AF) during hemodialysis. The system aims to provide unobtrusive and reliable detection using a specialized low power device and embedded algorithms. Future work includes optimizing device placement and incorporating additional signal quality metrics.

1. Photoplethysmography-based
system for atrial fibrillation
detection during hemodialysis
Dainius Stankevicius, A. Petrenas, A. Solosenko, M. Grigutis, T.
Januskevicius, L. Rimsevicius, and V. Marozas
dainius.stankevicius@ktu.lt
Vilnius university hospital
SANTARIŠKIŲ KLINIKOS

2. Introduction (1)
Hemodialysis:
– 350 000 people in the EU
– 5400 dialysis centers in the EU
– 60 % of chronic kidney disease patients
– 3 times per week
– 4 hours long
1/20

3. Introduction (2)
Atrial fibrillation (AF):
– The most common arrhythmia
– 33 million people around the world
– Risk of stroke and heart failure
Connection:
– Hemodialysis increases risk of developing AF
– AF occurs during hemodialysis
– Related to changes in fluid balance
2/20

4. Research question
Can we detect AF during hemodialysis in an
unobtrusive and still reliable way?
3/20

5. Proposal
• Electrocardiography is encumbering
• Photoplethysmography might be a solution
– Definitely less obtrusive!
– Also less reliable...
• A system consisting of:
– Specialized low power device
– Embedded algorithm to run the detection
4/20

6. We have built a device
• ARM Cortex-M0, 16 MHz, 16 kB SRAM
• PPG sampling at 250 Hz
• 24-hour battery life
5/20

7. Methodology: Start
ppg(n)
6/20

8. Raw PPG
7/20

9. Methodology: Preprocessing
ppg(n) Low pass filter
fc
LMS
w1...w10
m
+
-
x(n) = 1
ppgl(n)
ppga(n)
y(n)
8/20

10. Raw & Preprocessed PPGs
9/20

11. Methodology: SSF
SSF
N
ppga(n) s(n)
10/20
Slope Sum Function (SSF):

12. PPG and SSF
11/20

13. Methodology: Thresholding
SSF
N
Thresholding
T, Tk
max{s(n)}
arg[max{s(n)}]
median{p(m),...,p(m-4)}
i(m)-i(m-1)
ppga(n) s(n)
T
p(m)
i(m)
∆i(m)
st(n)SSF
N
ppga(n) s(n)
12/20

14. Thresholding and peak-to-peak
13/20

15. Methodology: AF detector
i(m)-i(m-1)
Low complexity
AF detector
a
d
t
i(m)
∆i(m) O(m)
14/20

16. Methodology: overall
Low pass filter
fc
LMS
w1...w10
m
SSF
N
Thresholding
T, Tk
max{s(n)}
arg[max{s(n)}]
median{p(m),...,p(m-4)}
i(m)-i(m-1)
+
-
Low complexity
AF detector
a
d
t
x(n) = 1
ppg(n) ppgl(n)
ppga(n)
s(n)
T
p(m)
i(m)
∆i(m) O(m)
st(n)
y(n)
15/20

17. Testing: hardware-in-the-loop
Real environment
Simulated environment
Pre-recorded PPG with
AF
Raw data file saved in
microSD card
Normal PPG
sampling routine...
Read PPG sample PPG
hardware
16/20

18. Results
0 50 100 150 200
-1
0
1
2
0 50 100 150 200
100
200
300
400
0 50 100 150 200
0
1
2
0 50 100 150 200
0
1
Outa
Outb Threshold
Time, s
0 50 100 150 200
0
5000
17/20

19. Conclusion
The proposed system has a potential to be
applied for an unobtrusive real-time AF
detection using solely the
photoplethysmogram.
18/20

20. Future directions
• Introduce signal quality metrics
• Introduce morphology-based features
• Determine the optimal placement of the
device
19/20

21. Acknowledgment
• This work was supported by CARRE
(No.611140) project, funded by the
European Commission Framework
Program 7.
• CARRE – CARdio REnal disease
• http://www.carre-project.eu
20/20